Optimized two-step (OTS) chemistry model for the description of partially premixed combustion

Abstract

Once properly optimized, single-step chemistry models are able to recover essential characteristics of flames such as burned gases temperature, flame propagation velocity and thickness. Ignition delay values can be reproduced as well. However, an improved description of either the premixed flame inner structure or the ignition process requires the consideration of the multi-step nature of chemical kinetics. Indeed, single-step chemistry models can neither describe properly the internal structure of laminar premixed flames nor the development of ignition processes. These limitations arise because single-step chemistry does not take into account intermediate species, the importance of which is essential. Thus, the present work is aimed at extending the recently-proposed optimized single-step (OSS) framework to multi-step chemistry. To this purpose an unbranched chain reaction with only two consecutive steps relevant to chain initiation and chain termination is considered. The corresponding kinetics model does involve two fictive species, the characteristics of which are optimized together with the two-step chemical scheme parameters, e.g., pre-exponential factors and activation energies. One of this fictive species is relevant to combustion products while the other represents the whole pool of radicals. The chemistry optimization procedure makes use of a well-defined in-house genetic algorithm and the resulting optimized two-step (OTS) model is assessed through comparisons made with data obtained from detailed chemistry computations used as reference. For similar (and even lower) CPU costs, the computations performed with the OTS model put into evidence significant improvements compared to the results obtained with the OSS description.